Abstract
This paper proposes a paradigm shift to the problem of infrastructure asset management modelling by focusing towards forecasting the future condition of the assets instead of using empirical modelling approaches based on historical data. The proposed prognostics methodology is general but, in this paper, it is applied to the particular problem of railway track geometry deterioration due to its important implications in the safety and the maintenance costs of the overall infrastructure. As a key contribution, a knowledge-based prognostics approach is developed by fusing on-line data for track settlement with a physics-based model for track degradation within a filtering-based prognostics algorithm. The suitability of the proposed methodology is demonstrated and discussed in a case study using published data taken from a laboratory simulation of railway track settlement under cyclic loads, carried out at the University of Nottingham (UK). The results show that the proposed methodology is able to provide accurate predictions of the remaining useful life of the system after a model training period of about 10% of the process lifespan.
Highlights
In most developed countries, the continuous ageing and the growing demand of use of critical infrastructures calls for advanced Prognostics and Health Management (PHM) concepts for optimal infrastructure asset management [1]
A knowledge-based prognostics methodology for railway track asset management has been developed in this paper
The proposed methodology is general but in this paper it was applied to the railway track due to its impact on the safety and the maintenance cost of the overall infrastructure
Summary
The continuous ageing and the growing demand of use of critical infrastructures calls for advanced Prognostics and Health Management (PHM) concepts for optimal infrastructure asset management [1]. Reliability Engineering and System Safety 181 (2019) 127–141 of the track, or, alternatively, using a purely deterministic physicsbased approach, a knowledge-based prognostics framework is proposed This approach fuses information from physics-based models and available data about track degradation within a Bayesian learning paradigm to sequentially reduce the initial modelling uncertainty [17] as long as new data are collected, so as to obtain increasingly accurate forecasts of the future condition of the track. The results show that the proposed prognostics methodology is able to accurately anticipate the future states of deformation of the track after a training period of about 10% of the total length of the process, which corresponds to the time required by the model to assimilate the data This prognostics performance is compared with the performance obtained for the same dataset using an empirical logarithmic model for track settlement.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
